mitch_de Posted November 20, 2010 Share Posted November 20, 2010 i am back! C2D E7500. (9* 333 MHz) LoadConfig: [hd(0,1)/Extra/com.apple.Boot.plist] 938 bytes. CPU: Intel® Core(tm)2 Duo CPU E7300 @ 2.66GHz max: 10 current: 6 CPU: Tjmax= ~ 104°C Temperature= ~ 45°C CPU: Vendor/Family/ExtFamily: 0x756e6547/0x6/0x0 CPU: Model/ExtModel/Stepping: 0x17/0x1/0x6 CPU: Multipliers x10: max=100, min=60 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: TSCFreq: 3330MHz CPU: CPUFreq: 3330MHz CPU: FSBFreq: 333MHz CPU: Enhanced SpeedStep: 1 CPU: NoCores/NoThreads: 2/2 CPU: Features: 0x268802a1 Chipset is G31 (pci id 0x29c08086) ..... Table #0 : Freeing backup [OK] Table #0 : Freeing [OK] Table #1 : Freeing backup [OK] Table #1 : Freeing [OK] Freeing map [OK] LoadConfig: [hd(0,1)/System/Library/CoreServices/SystemVersion.plist] 479 bytes. Loading Darwin 10.6 Loading kernel mach_kernel LoadDrivers: /Extra/Extensions.mkext LoadDrivers: /System/Library/Caches/com.apple.kext.caches/Startup/Extensions.mkext LoadACPI: [hd(0,1)/DSDT.aml] 11844 bytes. Using PCI-Root-UID value: 1 nVidia GeForce 9600 GT 512MB NV94 [10de:0622] :: PciRoot(0x1)/Pci(0x1,0x0)/Pci(0x0,0x0) NVCAP: 04000000-00000300-1c000000-0000000a-00000000 @0,display-cfg: ffffffff @1,display-cfg: ffffffff LoadConfig: [hd(0,1)/Extra/SMBIOS.plist] 526 bytes. Intel P35/G33 DRAM Controller [8086:29c0] at 00:00.0 Frequency detected: 532 MHz (1065) Dual Channel CAS:5 tRC:5 tRP:5 RAS:18 (5-5-5-18) Slot: 1 Type 19 2048MB (DDR2 SDRAM) 1066MHz Vendor=Kingston PartNo=2G-UDIMM SerialNo=BACCBB70 Slot: 3 Type 19 2048MB (DDR2 SDRAM) 1066MHz Vendor=Kingston PartNo=2G-UDIMM SerialNo=BACCC370 Patched DMI Table Found SMBIOS System Information Table 1 Customizing SystemID with : 00000000-0000-0000-0000-001d7dd11bf3 LoadACPI: [hd(0,1)/DSDT.aml] 11844 bytes. FADT: Restart Fix applied ! Found ACPI CPU: CPU0 Found ACPI CPU: CPU1 Found ACPI CPU: CPU2 Found ACPI CPU: CPU3 Voltage: min= 1037mV, max= 1112mV P-State: Starting from state P1 P-State: Added [Frequency 2997 MHz, FID 0x9, VID 0xea] P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0xe8] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0xe7] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0xe5] Patched ACPI version 1 DSDT Starting Darwin x86_64 PRO: The PStart key (new) works. IF i set 1, then it does PSate1(9*) ...PSate3(6*) PROBLEMS: 1. Even if i set butratio=9 (my BIOS setting, FSB is detected right as 333 MHz) the System (all MHZ Tools) showing 3333 MHZ as if busratio would be 10*, should be 3000 Mhz (9*333) CPU: Multipliers x10: max=100, min=60 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: TSCFreq: 3330MHz // wrong, busratio set to 9 CPU: CPUFreq: 3330MHz CPU: FSBFreq: 333MHz // OK Pstates generated in this case as 3000 MHz (9*) down to 1999 MHZ(6*) = OK! PStart=1 was computed OK, which (avoids 10*) P-State: Starting from state P1 P-State: Added [Frequency 2997 MHz, FID 0x9, VID 0xea] P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0xe8] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0xe7] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0xe5] 2. The moreVolt key, i set to 40 (40mV more each Pstate) seems to produce unusable values (since it was implemented) Normaly they were 1x...2x (hex), i get E5 to EA (sure not OK!) 3. maxVolt key seems have no effect ? Voltage: min= 1037mV, max= 1112mV / i set it to 1248mV 4. What does this mean ? Found ACPI CPU: CPU0 Found ACPI CPU: CPU1 Found ACPI CPU: CPU2 Found ACPI CPU: CPU3 I only have C2D , no Quad CPU ?! Also the OS X boottext shows only 2 CPUs enabled. PICS: 1. PSate=1, maxVolt=40, busratio=9 (VIDs bad), PStart worked) 2. busratio=9 (VIDs bad) 3. normal RC5 autogen (VIDs OK) 4. my .plist Link to comment Share on other sites More sharing options...
mitch_de Posted November 20, 2010 Share Posted November 20, 2010 Even if i disabled PStates gen in 5.0.5 + using handmade dsdt, the CPU freq stays wrong (FSB OK detected, busratio=9) back to RC5 boot, Pstate gen OFF, handmade dsdt Pstates, leaved all other (for RC5 unknown new keys unchanged) Now CPU freq is OK - no changes in the settings!!!! bdmesg RC5 msr(221): ia32_perf_stat 0x06000920 max: 10 current: 9 CPU: Vendor/Model/ExtModel: 0x756e6547/0x17/0x1 CPU: Family/ExtFamily: 0x6/0x0 CPU: MaxCoef/CurrCoef: 0xa/0x9 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: TSCFreq: 3330MHz CPU: FSBFreq: 333MHz CPU: CPUFreq: 2997MHz // OK CPU: NoCores/NoThreads: 2/2 CPU: Features: 0x0000025f Same settings 5.0.5: max: 10 current: 6 / ? CPU: Tjmax= ~ 104°C Temperature= ~ 45°C CPU: Vendor/Family/ExtFamily: 0x756e6547/0x6/0x0 CPU: Model/ExtModel/Stepping: 0x17/0x1/0x6 CPU: Multipliers x10: max=100, min=60 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: TSCFreq: 3330MHz CPU: CPUFreq: 3330MHz // WRONG CPU: FSBFreq: 333MHz CPU: Enhanced SpeedStep: 1 CPU: NoCores/NoThreads: 2/2 CPU: Features: 0x268802a1 That CPU freq problem must be new, i remember that wasnt using 5.0.1 or 5.0.2 (using also busratio=9). Q: Is the value of CPUFreq that value which changes by Pstate working ? I ask because if yes, we can use sysctl hw.cpufrequency to show that without need other tools (MSR didnt work for me, Systemprofiler shows always the other Cpu max.Freq (fixed max)). Link to comment Share on other sites More sharing options...
edgar87 Posted November 20, 2010 Share Posted November 20, 2010 With the latest nightly of Chameleon I get a score of 8400 in geekbench, and with the latest version of your bootloader I get a score of 8000.. With the system in my sig. EDIT: And with P-States generation enabled I only get a score of 4000 in geekbench Link to comment Share on other sites More sharing options...
Carstiman Posted November 20, 2010 Share Posted November 20, 2010 "i OC to 3840 Mhz only with FSB 333 and vcore to 1296 mV is there a easy way for me to get the pstate steps working in OC mode ?" Yep, valv does some major steps for the OC+ autogen Pstate customers! For my knowledge wie must use an combination of maxVolt= , to make higher CPU Volt possible and moreVolt= which does add some mVolts to each Pstate (to reach same mVolt you had before with handmade DSDT Pstates). valv also added some usage examples to each key - much thanks! Other people, which do also OC but also change busratio beside FSB may also need to use busratio= and PStart= (NEW in 5.0.4+) to avoid using the non available Pstate0, if busratio was set one step down (10>9) . I use OC : 9*333 (normal 10*266) so i had to use also PStart=1 to avoid PState0=10* autogeneration (if not KP because 3333 Mhz - to much for my 2666 MHz C2D). I will try that now with the 5.0.5 Version / Gigabyte EP35-DS3 / C2D E7500OC and report! UNCLEAR: I read in some other postings that the VID Problem also depends from the BIOS/Mainboard vendor and handling of CPU Volts settting (auto/normal vs. more Volts in BIOS). Most OCer will have set higher VID in the BIOS for OC. It can be, that at least at boottime (before bootloader+OS X sets up OCed VIDs) its needed to overVID in BIOS. Also it can be that the BIOS overVID does block / disallow any VID changes by the Pstates VID changes later. Would be nice if some coder here can tell something about that To all i cut out the OC related keys from the new helpfile.txt (within the .zip) GeneratePStates=Yes Automatically generate supported CPU P-States (default: disabled). Overclocking Options: (To use with CAUTION!) -------------------------------------------- PStates=<value> Maximum number of P-States to generate (default: all); e.g.: 8 to generate the 8 highest Frequency states only. PStart=<value> The highest Frequency state to start from (default: 0); e.g.: 1 to start from P1. busratio=<value> Alter bus ratio - e.g.: 8.5 (default: auto). minVolt=<value> Under-volt the processor to the specified value (mVolt); e.g.: 1100 for 1.1 V maxVolt=<value> Over-volt the processor to the specified value (mVolt); e.g.: 750 for 0.75 V moreVolt=<value> how much mVolts to add to each step; e.g.: 20 for getting additional 0.02 V for each CPU state lessVolt=<value> how much mVolts to take-out from each step; e.g.: 10 to substract 0.0.1 V from each CPU state tried your settings and played around with it. i lost one P-State: P-State: Added [Frequency 3830 MHz, FID 0x4b, VID 0x25] C-State: Adding 3 states: C1 C2 C3 Voltage: min= 1087mV, max= 1275mV P-State: Starting from state P1 P-State: Added [Frequency 3663 MHz, FID 0xb, VID 0xf4] P-State: Added [Frequency 3497 MHz, FID 0x4a, VID 0xf3] P-State: Added [Frequency 3330 MHz, FID 0xa, VID 0xf2] P-State: Added [Frequency 3164 MHz, FID 0x49, VID 0xf1] P-State: Added [Frequency 2997 MHz, FID 0x9, VID 0xef] P-State: Added [Frequency 2831 MHz, FID 0x48, VID 0xee] P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0xed] P-State: Added [Frequency 2498 MHz, FID 0x47, VID 0xec] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0xea] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0xe9] but i think first i need support for maxVolt= support for 1325 mV Link to comment Share on other sites More sharing options...
Time2Retire Posted November 20, 2010 Share Posted November 20, 2010 With the latest nightly of Chameleon I get a score of 8400 in geekbench, and with the latest version of your bootloader I get a score of 8000.. With the system in my sig. EDIT: And with P-States generation enabled I only get a score of 4000 in geekbench Same here. Also with OS X 10.6.5 I presume? Link to comment Share on other sites More sharing options...
edgar87 Posted November 20, 2010 Share Posted November 20, 2010 OS X 10.6.5 I presume? Yep, 10.6.5, why? This is my boot.plist: com.apple.boot.plist.zip Link to comment Share on other sites More sharing options...
Time2Retire Posted November 20, 2010 Share Posted November 20, 2010 Yep, 10.6.5, why? This is my boot.plist: Apple may have changed their OSPM code for (certain) Intel Core iN CPU's. Went back to OS X 10.6.4 to check (with the same booter and P_States) and everything is fine again. Too bad that MSR Tools only shows two cores (and seem to be stuck at the second P_State) and that VoodooMonitor triggers a panic on my notebook. I wish there was a simple command line tool to generate P-States. This way people could test things, but without having to replace the boot file. Which in my opinion is a rather dated requirement. Link to comment Share on other sites More sharing options...
MiniHack Posted November 20, 2010 Share Posted November 20, 2010 A quick question that I hope you can answer. My output of BDMESG is below using an experimental DSDT on my Vostro 3500. The key thing in this is that with this bootloader I have for the first time been able to generate p-states on my laptop despite my inability for me to remove the NullCPUPowerManagement.kext. My question is though will these p-states be used if the disabler is still in place and how can I check this? Versions of CPU-X and voodoomonitor do not seem to work on the i5 CPUs so I'd appreciate a comment from someone who knows more than me! BTW removing the disabler kext has been tried by me many times and my DSDTs have been much edited to try and enable that over the last months, so if AnVal is a genuine way to get speedstep working then this is real progression.... Thanks for your comments. CPU: Intel(R) Core(TM) i5 CPU M 430 @ 2.27GHz CPU: Tjmax= ~ 105°C Temperature= ~ 53°C CPU: Vendor/Family/ExtFamily: 0x756e6547/0x6/0x0 CPU: Model/ExtModel/Stepping: 0x25/0x2/0x2 CPU: MaxCoef/CurrCoef: 0x0/0x11 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: Multiplier x10: 90 CPU: TSCFreq: 2261MHz CPU: CPUFreq: 1197MHz CPU: FSBFreq: 133MHz CPU: Enhanced SpeedStep: 1 CPU: NoCores/NoThreads: 8/16 CPU: Features: 0x369802a1 Chipset is UNKNOWN (pci id 0x448086) We have an NVIDIA Card .... Found ACPI CPU: CPU0 Found ACPI CPU: CPU1 Found ACPI CPU: CPU2 Found ACPI CPU: CPU3 Found ACPI CPU: CPU4 Found ACPI CPU: CPU5 Found ACPI CPU: CPU6 Found ACPI CPU: CPU7 Voltage: min= 700mV, max= 700mV P-State: Added [Frequency 1197 MHz, FID 0x9, VID 0x0] P-State: Added [Frequency 1064 MHz, FID 0x8, VID 0x0] P-State: Added [Frequency 931 MHz, FID 0x7, VID 0x0] P-State: Added [Frequency 798 MHz, FID 0x6, VID 0x0] .. Link to comment Share on other sites More sharing options...
rednous Posted November 20, 2010 Share Posted November 20, 2010 valv, great to see you back here and the newer booter, too. My bdmesg log from AnVaL 5.0.5 is following below. I like very much the updated BootHelp.txt. I don't over/ under clock my PC and respectively don't use these keys in the com.apple.boot.plist. Should you need any specific test to perform and report just give me the directions. LoadConfig: [hd(0,2)/Extra/com.apple.Boot.plist] 1054 bytes. CPU: Intel® Core(tm)2 Duo CPU E7500 @ 2.93GHz max: 11 current: 6 CPU: Tjmax= ~ 105°C Temperature= ~ 44°C CPU: Vendor/Family/ExtFamily: 0x756e6547/0x6/0x0 CPU: Model/ExtModel/Stepping: 0x17/0x1/0xa CPU: Multipliers x10: max=110, min=60 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: TSCFreq: 2933MHz CPU: CPUFreq: 2933MHz CPU: FSBFreq: 266MHz CPU: Enhanced SpeedStep: 1 CPU: NoCores/NoThreads: 2/2 CPU: Features: 0x268802a1 Chipset is P45 (pci id 0x2e208086) We have an NVIDIA Card Creating 2 Mode Tables New table with id : 0 New table with id : 1 First Standard VESA Table at offset 0x7fe3 Second Standard VESA Table at offset 0x4b9f Table #0 has 16 modes Table #1 has 33 modes Aspect Ratio is 16/10 Patching Table #0 : Mode 1024x768 -> 1920x1200 (1047 1183 1343 770 776 805) Patching Table #1 : Mode 640x200 -> 640x400 (664 672 720 408 H- V+) Mode 640x350 -> 640x400 (664 678 720 408 H- V+) Mode 640x480 -> 640x400 (669 685 720 408 H- V+) Mode 640x480 -> 640x400 (669 685 720 408 H- V+) Mode 800x600 -> 800x500 (836 852 900 512 H- V+) Mode 800x600 -> 800x500 (836 852 900 512 H- V+) Mode 1024x768 -> 1024x640 (1071 1087 1152 656 H- V+) Mode 1024x768 -> 1024x640 (1071 1087 1152 656 H- V+) Mode 1280x1024 -> 1280x800 (1343 1359 1440 823 H- V+) Mode 1280x1024 -> 1280x800 (1343 1359 1440 823 H- V+) Mode 1280x720 -> 1280x800 (1324 1340 1440 807 H- V-) Mode 1400x1050 -> 1400x875 (1474 1488 1575 908 H+ V+) Mode 1600x1200 -> 1600x1000 (1674 1690 1800 1025 H- V+) Mode 2048x1536 -> 2048x1280 (2144 2160 2048 1314 H+ V+) LoadConfig: [hd(0,2)/Extra/Themes/mac/theme.plist] 2799 bytes. Resolution : 1920x1200 (Graphics Mode key) LoadConfig: [hd(0,2)/Extra/Themes/mac/theme.plist] 2799 bytes. Found mode 1920x1200 in VESA Table LoadConfig: [hd(0,2)/Extra/com.apple.Boot.plist] 1054 bytes. Setting boot-uuid to: 43B13315-6763-38BB-A562-E1EF385BA202 Closing VBios Table #0 : Freeing backup [OK] Table #0 : Freeing [OK] Table #1 : Freeing backup [OK] Table #1 : Freeing [OK] Freeing map [OK] LoadConfig: [hd(0,2)/System/Library/CoreServices/SystemVersion.plist] 479 bytes. Loading Darwin 10.6 Loading kernel mach_kernel LoadConfig: [hd(0,2)/Extra/Extensions/FakeSMC.kext/Contents/Info.plist] 3727 bytes. LoadConfig: [hd(0,2)/Extra/Extensions/IntelThermal.kext/Contents/Info.plist] 1501 bytes. LoadConfig: [hd(0,2)/Extra/Extensions/OrangeIconFix.kext/Contents/Info.plist] 1709 bytes. LoadDrivers: /System/Library/Caches/com.apple.kext.caches/Startup/Extensions.mkext LoadACPI: [hd(0,2)/Extra/DSDT.aml] 3747 bytes. Using PCI-Root-UID value: 0 nVidia GeForce 9800 GT 512MB NV92 [10de:0614] :: PciRoot(0x0)/Pci(0x1,0x0)/Pci(0x0,0x0) NVCAP: 04000000-00000300-1c000000-0000000a-00000000 @0,display-cfg: ffffffff @1,display-cfg: ffffffff LAN Controller [10ec:8168] :: PciRoot(0x0)/Pci(0x1c,0x5)/Pci(0x0,0x0) Setting up lan keys LoadConfig: [hd(0,2)/Extra/SMBIOS.plist] 661 bytes. Intel P45/G45 DRAM Controller [8086:2e20] at 00:00.0 Frequency detected: 400 MHz (800) Dual Channel CAS:4 tRC:4 tRP:4 RAS:12 (4-4-4-12) Slot: 0 Type 19 2048MB (DDR2 SDRAM) 800MHz Vendor=G Skill Intl PartNo=F2-6400CL4-2GBPK SerialNo=00000000 Slot: 1 Type 19 2048MB (DDR2 SDRAM) 800MHz Vendor=G Skill Intl PartNo=F2-6400CL4-2GBPK SerialNo=00000000 Slot: 2 Type 19 2048MB (DDR2 SDRAM) 800MHz Vendor=Mushkin PartNo=991580996580 SerialNo=00000000 Slot: 3 Type 19 2048MB (DDR2 SDRAM) 800MHz Vendor=Mushkin PartNo=991580996580 SerialNo=00000000 Patched DMI Table Found SMBIOS System Information Table 1 Customizing SystemID with : 00000000-0000-0000-0000-001fd0928ace LoadACPI: [hd(0,2)/Extra/DSDT.aml] 3747 bytes. Converted ACPI RSD PTR version 1 to version 3 Converted RSDT table to XSDT table Converted ACPI V1 FADT to ACPI V4 FADT FADT: Restart Fix applied ! Found ACPI CPU: CPU0 Found ACPI CPU: CPU1 C-State: Adding 5 states: C1 C2 C3 C4 C6 Voltage: min= 987mV, max= 1150mV P-State: Added [Frequency 2926 MHz, FID 0xb, VID 0x23] P-State: Added [Frequency 2660 MHz, FID 0xa, VID 0x21] P-State: Added [Frequency 2394 MHz, FID 0x9, VID 0x1f] P-State: Added [Frequency 2128 MHz, FID 0x8, VID 0x1c] P-State: Added [Frequency 1862 MHz, FID 0x7, VID 0x1a] P-State: Added [Frequency 1596 MHz, FID 0x6, VID 0x17] Converted ACPI V1 FADT to ACPI V4 FADT FADT: Restart Fix applied ! Patched ACPI version 2 DSDT Starting Darwin x86_64 Link to comment Share on other sites More sharing options...
mitch_de Posted November 20, 2010 Share Posted November 20, 2010 i lost one P-State: P-State: Added [Frequency 3830 MHz, FID 0x4b, VID 0x25] C-State: Adding 3 states: C1 C2 C3 Voltage: min= 1087mV, max= 1275mV P-State: Starting from state P1 // thats because PStart=1, working! P-State: Added [Frequency 3663 MHz, FID 0xb, VID 0xf4] P-State: Added [Frequency 3497 MHz, FID 0x4a, VID 0xf3] P-State: Added [Frequency 3330 MHz, FID 0xa, VID 0xf2] P-State: Added [Frequency 3164 MHz, FID 0x49, VID 0xf1] P-State: Added [Frequency 2997 MHz, FID 0x9, VID 0xef] P-State: Added [Frequency 2831 MHz, FID 0x48, VID 0xee] P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0xed] P-State: Added [Frequency 2498 MHz, FID 0x47, VID 0xec] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0xea] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0xe9] but i think first i need support for maxVolt= support for 1325 mV Yep, that because PStart=1 is working already - that disables Pstate0 (max. multi). Thats OK so. You dont need PStart= if you dont have limited the multiplier in BIOS (like me : 9 vs 10 possible). I can see also, that your VIDs , using moreVolt key got also wrong (Ex,Fx values insted of 1x,2x values). That wromg values are so high that - the good - they are not used my the cpu (over any limit). maxVolt seems also to be not used, does nothing . I dont know if there is some debug code which would show that, beside the min/max Volt informatons shown in bdmesg We must wait for valv, because its sure complex work to do all that for complete different cpus! Link to comment Share on other sites More sharing options...
Time2Retire Posted November 20, 2010 Share Posted November 20, 2010 The PerformanceStateArray has change in OS X 10.6.5. Here's the one from OS X 10.6.4 <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 da[/color] 00 00 [color="#FF0000"]00 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 55[/color] 00 00 [color="#FF0000"]00 01 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 cf[/color] 00 00 [color="#FF0000"]00 02 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 4a[/color] 00 00 [color="#FF0000"]00 03 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 c5[/color] 00 00 [color="#FF0000"]00 04 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 3f[/color] 00 00 [color="#FF0000"]00 05 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 ba[/color] 00 00 [color="#FF0000"]00 06 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 35[/color] 00 00 [color="#FF0000"]00 07 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 af[/color] 00 00 [color="#FF0000"]00 08 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 2a[/color] 00 00 [color="#FF0000"]00 09 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]00 a5[/color] 00 00 [color="#FF0000"]00 0a 00 00[/color]> And here's the one from OS X 10.6.5 <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]da 08[/color] 00 00 [color="#FF0000"]00 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]55 08[/color] 00 00 [color="#FF0000"]01 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]cf 07[/color] 00 00 [color="#FF0000"]02 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]4a 07[/color] 00 00 [color="#FF0000"]03 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]c5 06[/color] 00 00 [color="#FF0000"]04 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]3f 06[/color] 00 00 [color="#FF0000"]05 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]ba 05[/color] 00 00 [color="#FF0000"]06 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]35 05[/color] 00 00 [color="#FF0000"]07 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]af 04[/color] 00 00 [color="#FF0000"]08 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]2a 04[/color] 00 00 [color="#FF0000"]09 00 00 00[/color]> <00 00 00 00 00 00 00 00 10 00 00 00 10 00 00 00 [color="#FF0000"]a5 03[/color] 00 00 [color="#FF0000"]0a 00 00 00[/color]> Same P-States! Note: Apple clearly fixed a previously made programing error. Do you spot the error? I do Link to comment Share on other sites More sharing options...
valv Posted November 20, 2010 Author Share Posted November 20, 2010 This one is very experimental: Should be fixing all (or most) of the above behaviors. I 'd need special feedback from guyz with i-series cpus about turbo. From my other friends, I 'd appreciate to know if voltage, multiplier... issues are fixed now. Thanks to all, and sorry for the mess @dutchhokeypro, now that's a find. thank u for sharing a previously made programing error.I 'd add that this may not be a generic error for all models, though Link to comment Share on other sites More sharing options...
mitch_de Posted November 20, 2010 Share Posted November 20, 2010 Here we go (with boot_4) The good: CPU Mhz now ok, 3000MHz (9*333 max), before 3333 Mhz CPU: TSCFreq: 3330MHz CPU: CPUFreq: 2997MHz // OK now CPU: FSBFreq: 333MHz The bad: 1. Min max Volts extreme high , and Min Volt Voltage: min= 3575mV, max= 3112mV (before the min/max looked normal, MaxVolt had no effect) 2. Sticking with [FSB: 0Mhz, Bus-Ratio: 9] // 0 Mhz , perhaps some sign of bug for later computation? Bus-Ratio: min=6, max=10 // CPU can this: yes, but busratio=9 set and also limited in BIOS=9, but perhaps later computed OK=dosnt matter 3. Now, also with PStart=1 (disables PState0) i get now only 3 Pstates , before i get 4 (PSstart=1) or all 5 (Pstart=0) Now 8*,7*,6* are generated (missed 9*, 10* is disabled by PStart=1) Can there be an bug / interfernence between PStart=1 and also set busratio=9 ? Perhaps programm takes now (newest) busratio=9 and disables that 9* , because it thinks it is PState0 , but Pstate0 is 10* ? 4. VIDs now not Ex anymore but also much to high (normal range 12...25 hex). /Extra/bdmesg ; exit; CPU: Intel® Core2 Duo CPU E7300 @ 2.66GHz Bus-Ratio: min=6, max=10 Sticking with [FSB: 0Mhz, Bus-Ratio: 9] // 0 Mhz ? CPU: Tjmax= ~ 104°C Temperature= ~ 41°C CPU: Vendor/Family/ExtFamily: 0x756e6547/0x6/0x0 CPU: Model/ExtModel/Stepping: 0x17/0x1/0x6 CPU: Multipliers x10: max=90, min=60 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: TSCFreq: 3330MHz CPU: CPUFreq: 2997MHz CPU: FSBFreq: 333MHz CPU: Enhanced SpeedStep: 1 CPU: NoCores/NoThreads: 2/2 CPU: Features: 0x268802a1 Chipset is G31 (pci id 0x29c08086) We have an NVIDIA Card Creating 2 Mode Tables ... ... Voltage: min= 3575mV, max= 3112mV P-State: Starting from state P1 // disabled 9* and 10*, 10* s OK, 9* shouldnt be disabled by PStart=1 and busratio=9 P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0xbd] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0xb8] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0xb3] Patched ACPI version 1 DSDT Starting Darwin x86_64 Link to comment Share on other sites More sharing options...
Carstiman Posted November 20, 2010 Share Posted November 20, 2010 Here we go (with boot_4) The good: CPU Mhz now ok, 3000MHz (9*333 max), before 3333 Mhz CPU: TSCFreq: 3330MHz CPU: CPUFreq: 2997MHz // OK now CPU: FSBFreq: 333MHz The bad: 1. Min max Volts extreme high , and Min Volt < max Volt, i use MaxVolt=1248 Voltage: min= 3575mV, max= 3112mV (before the min/max looked normal, MaxVolt had no effect) 2. Sticking with [FSB: 0Mhz, Bus-Ratio: 9] // 0 Mhz , perhaps some sign of bug for later computation? Bus-Ratio: min=6, max=10 // CPU can this: yes, but busratio=9 set and also limited in BIOS=9, but perhaps later computed OK=dosnt matter 3. Now, also with PStart=1 (disables PState0) i get now only 3 Pstates , before i get 4 (PSstart=1) or all 5 (Pstart=0) Now 8*,7*,6* are generated (missed 9*, 10* is disabled by PStart=1) Can there be an bug / interfernence between PStart=1 and also set busratio=9 ? Perhaps programm takes now (newest) busratio=9 and disables that 9* , because it thinks it is PState0 , but Pstate0 is 10* ? 4. VIDs now not Ex anymore but also much to high (normal range 12...25 hex). /Extra/bdmesg ; exit; CPU: Intel® Core™2 Duo CPU E7300 @ 2.66GHz Bus-Ratio: min=6, max=10 Sticking with [FSB: 0Mhz, Bus-Ratio: 9] // 0 Mhz ? CPU: Tjmax= ~ 104°C Temperature= ~ 41°C CPU: Vendor/Family/ExtFamily: 0x756e6547/0x6/0x0 CPU: Model/ExtModel/Stepping: 0x17/0x1/0x6 CPU: Multipliers x10: max=90, min=60 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: TSCFreq: 3330MHz CPU: CPUFreq: 2997MHz CPU: FSBFreq: 333MHz CPU: Enhanced SpeedStep: 1 CPU: NoCores/NoThreads: 2/2 CPU: Features: 0x268802a1 Chipset is G31 (pci id 0x29c08086) We have an NVIDIA Card Creating 2 Mode Tables ... ... Voltage: min= 3575mV, max= 3112mV P-State: Starting from state P1 // disabled 9* and 10*, 10* s OK, 9* shouldnt be disabled by PStart=1 and busratio=9 P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0xbd] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0xb8] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0xb3] Patched ACPI version 1 DSDT Starting Darwin x86_64 same too me. boot4.txt Link to comment Share on other sites More sharing options...
Time2Retire Posted November 20, 2010 Share Posted November 20, 2010 @dutchhokeypro, now that's a find. thank u for sharingI 'd add that this may not be a generic error for all models, though Yeah. OS X 10.6.4 was apparently much more forgiving, but the slightest error now – like a wrong multiplier – is now punished with a much lower Geekbench score. I had 4033 with 10.6.4 and that dropped to a low 1998 after the upgrade. I solved this by applying the correct multipliers and I am back to 4033 with 10.6.5 The problem was that my in U-EFI _PSS values where used. I had no clue. Your code however appear to be fine. Just to make things clear. Link to comment Share on other sites More sharing options...
Maniac10 Posted November 20, 2010 Share Posted November 20, 2010 Here's my log, the only way to boot is to disable GE and run bdmesg in single user mode. I've booted with generate PS and CS, GE=no -s so you can take a look at the results for my i7. I can attach an ioreg dump and bdmesg with the kabyl boot to compare if you want. Link to comment Share on other sites More sharing options...
MiniHack Posted November 20, 2010 Share Posted November 20, 2010 Here is my BDMESG output using the 5.0.5 version on i5 430M. Note there are quite a few more p-states reported over the 5.0.3 version. C1 C3 C6 Voltage: min= 700mV, max= 700mV P-State: Added [Frequency 2508 MHz, FID 0x13, VID 0x0] P-State: Added [Frequency 2376 MHz, FID 0x12, VID 0x0] P-State: Added [Frequency 2244 MHz, FID 0x11, VID 0x0] P-State: Added [Frequency 2112 MHz, FID 0x10, VID 0x0] P-State: Added [Frequency 1980 MHz, FID 0xf, VID 0x0] P-State: Added [Frequency 1848 MHz, FID 0xe, VID 0x0] P-State: Added [Frequency 1716 MHz, FID 0xd, VID 0x0] P-State: Added [Frequency 1584 MHz, FID 0xc, VID 0x0] P-State: Added [Frequency 1452 MHz, FID 0xb, VID 0x0] P-State: Added [Frequency 1320 MHz, FID 0xa, VID 0x0] P-State: Added [Frequency 1188 MHz, FID 0x9, VID 0x0] P-State: [Frequency 1056 MHz, FID 0x8, VID 0x0] is over the limit. Removed! P-State: [Frequency 924 MHz, FID 0x7, VID 0x0] is over the limit. Removed! P-State: [Frequency 792 MHz, FID 0x6, VID 0x0] is over the limit. Removed! FADT: Restart Fix applied ! Patched ACPI version 2 DSDT Link to comment Share on other sites More sharing options...
valv Posted November 20, 2010 Author Share Posted November 20, 2010 The bad:1. Min max Volts extreme high , and Min Volt < max Volt That's not good.2. Sticking with [FSB: 0Mhz, Bus-Ratio: 9] // 0 Mhz , perhaps some sign of bug for later computation?Bus-Ratio: min=6, max=10 // CPU can this: yes, but busratio=9 set and also limited in BIOS=9, but perhaps later computed OK=dosnt matter this is cosmetic, but should be fixed3. Now, also with PStart=1 (disables PState0) i get now only 3 Pstates , before i get 4 (PSstart=1) or all 5 (Pstart=0)Now 8*,7*,6* are generated (missed 9*, 10* is disabled by PStart=1) Can there be an bug / interfernence between PStart=1 and also set busratio=9 ? Perhaps programm takes now (newest) busratio=9 and disables that 9* , because it thinks it is PState0 , but Pstate0 is 10* ? you need to use one or the other4. VIDs now not Ex anymore but also much to high (normal range 12...25 hex).this is a priority.ok, I 'll be back asap; for the moment am removing boot_4 Link to comment Share on other sites More sharing options...
kizwan Posted November 20, 2010 Share Posted November 20, 2010 @valv, This is bdmesg output on Dell Studio:- Studio1557_boot_4.txt CPU frequency reported in System Profiler is 1.6GHz (normal operating frequency). Link to comment Share on other sites More sharing options...
mitch_de Posted November 20, 2010 Share Posted November 20, 2010 "3. Now, also with PStart=1 (disables PState0) i get now only 3 Pstates , before i get 4 (PSstart=1) or all 5 (Pstart=0) Now 8*,7*,6* are generated (missed 9*, 10* is disabled by PStart=1) Can there be an bug / interfernence between PStart=1 and also set busratio=9 ? valv: you need to use one or the other // To others: busration also has function for Pstate autogen, disables highest Pstate if busratio YEP ! I now used PStart=0 with busratio=9 and all wanted PState were generated :censored2: P-State: Added [Frequency 2997 MHz, FID 0x9, VID 0xb8] P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0xb5] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0xb0] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0xab] Wreid : I set MoreVolt=0 (before 40) and also some ?? VIDs. I think that key must removed until fix, because =0 also makes not the normal(OK) VIDs generated without that key. Now, removed MaxVolt and the moreVolt keys, added C-State gen C-State: Adding 3 states: C1 C2 C3 Voltage: min= 1037mV, max= 1112mV // much diff to apply MaxVolt P-State: Added [Frequency 2997 MHz, FID 0x9, VID 0x20] // also removing MoreVolt key get back normal VIDs P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0x1f] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0x1d] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0x1b] Starting Darwin x86_64 What do you meaned that my "The good" : fixed MHZ shown isnt good ? CPU: Multipliers x10: max=90, min=60 CPU: MaxDiv/CurrDiv: 0x0/0x0 CPU: TSCFreq: 3330MHz CPU: CPUFreq: 2997MHz /GOOD, before was wrong 3333 CPU: FSBFreq: 333MHz Or do you mean with not good, that TSC freq - i dont know whats this is with 3333 Mhz? C-State: Adding 3 states: C1 C2 C3 I thought my C2D E7400 has only C1, not C2+C3 . And the default is that C2+ are disabled and must enabled by: EnableC2State=Yes Add C2 and lower supported states to _CST section (default: disabled). EnableC3State=Yes Add C3 and lower supported states to _CST section (default: disabled). Should i better add those keys with No to disable them active ? EDIT: I added for test EnableC3State=No, but also C1+C2+C3 were generated. So something must be wrong with the default disabled C2,C3,... . Looks that the major EnableCStates enables all - even the default disabled C2 and above) TO OTHERS: Try to cut out the long bdmesg infos or use code box to post. Uncut/non codebox bdmesg postings take much space / long scrolling/eye lost the essenials of that postings - THANKS Link to comment Share on other sites More sharing options...
valv Posted November 20, 2010 Author Share Posted November 20, 2010 first of, Big Thank you to all! @mitch_de, should the PStart key be removed now, as it makes redundancy with the bus-ratio injection? what do u think? regarding more/lessVolt keys, it could be fixed with boot_5 (untested!). c-states generation behaves differently. Note: EnableC4State=Yes (for example): c4 and all previous states (supported by your cpu) would be generated. If you use one of the EnableCXState keys, no need for the GenerateCStates key EnableC4State=Yes (but cpu doesn't support it): c4 would not be generated If u need to disable c3 (but keep the lower states): you need to remove EnableC3State altogether and add EnableC2State. @kizwan, sorry for the delay, but I honestly don't know what to say other than: "u have pretty nice machines. I'd hope I could afford it ". Oh! and thank you for testing Next attempt: boot5 is now useless Link to comment Share on other sites More sharing options...
mitch_de Posted November 20, 2010 Share Posted November 20, 2010 "mitch_de, should the PStart key be removed now, as it makes redundancy with the bus-ratio injection?" Yes, if busratio is working and also it does limit PState i would comment the usage of busratio with " also limit fastest PSate to busratio". Any hint, which way / tool can be used to show changing CPU MHz? MSR Tool dient work for me, and all others only show the max CPU MHz. Reboot with boot5 , EnableC2=Yes, added volt keys again.... I´LL BE BACK Link to comment Share on other sites More sharing options...
valv Posted November 20, 2010 Author Share Posted November 20, 2010 Any hint, which way / tool can be used to show changing CPU MHz? MSR Tool dient work for me, and all others only show the max CPU MHz.CoolBook !? Link to comment Share on other sites More sharing options...
mitch_de Posted November 20, 2010 Share Posted November 20, 2010 NO - i also thougth that would be nice, but it uses own Pstates (and reads/shows them). You see i am BACK with boot5 and have some gift for you boot5, moreVolt=40 , MaxVolt=1248 (with Versions before gave 3.x Volts) C-State: Adding 3 states: C1 C2 C3 Voltage: min= 1037mV, max= 1237mV / without look my older postings, MaxVolt is WORKING NOW! P-State: Added [Frequency 2997 MHz, FID 0x9, VID 0x2d] // also VIDs working now if MoreVolt is used, Normal is 20hex P-State: Added [Frequency 2664 MHz, FID 0x8, VID 0x2a] P-State: Added [Frequency 2331 MHz, FID 0x7, VID 0x25] P-State: Added [Frequency 1998 MHz, FID 0x6, VID 0x20] Also (more cosmetic) fixed debug info: Sticking with [FSB: 333Mhz, Bus-Ratio: 9], before FSB: 0 Mhz Only i cant get rid of C3 State, even i use EnableC2State=Yes (should disable C3,C4,...) Link to comment Share on other sites More sharing options...
kizwan Posted November 20, 2010 Share Posted November 20, 2010 u have pretty nice machines. Thanks. It would be very nice if the GPU is supported in OSX. The Dell is not a mistake since I bought it for work reason. The Asus was for OSX to be honest but it is a mistake since I didn't know it have Optimus switchable graphic card (poor advertisement). But at least it have good GPU for gaming. Link to comment Share on other sites More sharing options...
Recommended Posts